01086nam a2200181 a 450000100080000000500110000800800410001902000180006010000160007824500610009426000590015530000110021452005920022565000200081765000330083765300180087070000160088820912972018-06-06       2001    bl uuuu      00u1 u    #d  a0-521-79298-31 aROBERTS, S.  aIndependent component analysisbprinciples and practice.  aCambridge; New York : Cambridge University Pressc2001  a338 p.  aIntroduction. Fast ICA by a fixed-point algorithm that maximizes non-gaussianity. ICA, graphical models and variational methods. Nonlinear. Separation of non-stationary natural signals. Separation of non-stationary sources: algorithm and performance. Blind source separation by sparse decomposition in asignal dictionary. Ensemble learning for blind source separation. Image processing methods using ICA mixture models. Latent class and trait models for data classification and visualisation. Particle filters for non-stationary ICA. ICA: model order selection and dynamic source models.  aNeural networks  aPrincipal component analysis  aRedes neurais1 aEVERSON, R.